Will Intel Foundry's FY2026 segment operating loss narrow year-over-year versus the FY2025 $(10.3)B loss?

The resolution bar is low — ANY narrowing from the $(10.3)B baseline counts as YES. The FY25 loss included a one-time $800M Q2 impairment plus $200M in period costs for excess prior-gen tools that will NOT repeat in FY26, creating ~$1B of automatic YoY improvement absent any operational change. With ~80-85% of the loss being fixed depreciation, management guiding revenue 'up double-digits QoQ' in Q1 2026 on EUV mix and 18A pricing, and CapEx 'flat to down slightly,' the mechanical arithmetic favors narrowing. The Q4 2025 $188M worse QoQ from 18A early ramp is a headwind but scales to ~$750M annualized if sustained — less than the one-time offset.

Mechanical tailwinds are real — the $1B non-repeating charges plus revenue growth from EUV/18A ramp create a strong base case for narrowing. However, I'm pricing in meaningful offsetting risk: the 18A ramp is explicitly described as a continuing drag ('Q4 $188M worse QoQ driven by 18A early ramp'), and this headwind runs through most of FY26. Panther Lake internal transfers are dilutive at lower effective margins, and committee flagged roadkill-radar's wider $4-6B cash burn framing as a legitimate alternative lens. Base rate for turnaround year-2 segment improvement given 'MEETING' execution is ~60-70%, and the specific one-time offset adds ~8-10pp.

I weight the accounting mechanics heavily. The FY25 baseline includes $1B of charges (Q2 $800M impairment + $200M excess tool period costs) that have no FY26 analog — that's a floor already ~$1B better before operational changes. Depreciation is largely fixed given multi-year tooling schedules and 'flat to down' CapEx. For the NO outcome, Intel would need operational deterioration of >$1B to offset the non-repeat — plausible only if 18A ramp drag compounds AND revenue fails AND Panther Lake dilution cascades. Given management execution track record (five revenue beats, OpEx on target, CapEx -39% YoY), operational execution argues for incremental improvement not deterioration.

This is a low-bar binary: any narrowing vs $(10.3)B resolves YES. The single most decisive fact is the FY25 $800M Q2 impairment plus $200M period costs that will not repeat — that's automatically ~$1B better FY26 without touching operations. Layer on guided Q1 2026 Foundry revenue up double-digits QoQ on EUV and 18A, DCAI's fastest sequential growth in a decade driving internal transfer demand, and management's demonstrated cost discipline. The risk is that 18A ramp drag sustained all year could eat through the $1B cushion, but that requires the drag to compound from the Q4 2025 $188M QoQ hit rather than normalize as yields improve monthly.

I'm more cautious than the mechanical read suggests. Yes, the $1B one-time won't repeat — but committee explicitly flagged that Q4 2025 was WORSENING not improving, and that 18A ramp is an ongoing drag. If Q4's $188M QoQ deterioration continues into FY26 (even at half-pace), you get $300-500M of incremental drag. Panther Lake dilution is real. The committee also flagged potential FY26 segment reclassifications that could mechanically move things either direction — a non-operational source of noise. Still favor YES because the bar is so low and the one-time is so clean, but the committee's conflicting-signal framing on foundry burn severity warrants a less confident estimate.

Reading the prediction context directly: the committee itself flags base rate for narrowing when prior year had one-time impairments that won't repeat AND execution is 'MEETING' AND capex is throttling as MODERATE-HIGH (~60-70%). I adjust upward from that base rate because two of the three conditions are already banked (impairment, capex) and operational execution has been demonstrated for five consecutive quarters. Revenue tailwind from EUV mix progression (<1% to >10%) and 18A pricing uplift provide a second lever. Primary downside is that the market resolves on REPORTED segment loss (not cash), and segment attribution volatility could move the headline independently.

Low bar: any FY26 loss better than $(10.3)B resolves YES. The $800M Q2 2025 impairment plus $200M period costs won't repeat, giving ~$1B automatic improvement. Management guides revenue up double-digits QoQ on EUV mix and 18A. Main risk is sustained 18A ramp drag. Favors YES.

Committee base rate for YoY segment loss narrowing under these conditions is 60-70%. Strong tailwinds: non-recurring $1B, CapEx flat-to-down, execution MEETING. Offsets: 18A ramp drag continuing, Panther Lake dilution. Q4 2025 trajectory was worsening which tempers confidence. Settle slightly above committee base rate.

The market resolves on reported segment operating loss vs $(10.3)B. The $1B one-time charges in FY25 baseline are the single most important fact — that's a structural floor of improvement independent of operations. Depreciation largely fixed. Revenue guided up. CapEx throttled. The only way this resolves NO is operational deterioration >$1B, which is hard to reconcile with MEETING execution.